I don't think this is so easy. While working on Doxia we have noticed
that a lot of intertwined and as soon as you touch on component you need
to touch the rest also.
At least a one time pain we must go.
Am 2022-03-13 um 12:52 schrieb Romain Manni-Bucau:
Hi,
No big issue with me but I would also consider letting v3 as it is and
start working on v4 with an alpha coming soon maybe?
Can be a way to reach these goals without impacting twice end users too.
Le dim. 13 mars 2022 à 12:48, Michael Osipov <micha...@apache.org> a écrit :
Folks,
as you might now we are dragging old luggage over and over with every
new year. We are making (little) progress with cleansing, skimming and
updating of our components.
I'll will make a radical proposal I have already discussed privately
with other fellow committers over Slack:
Raise the entire baseline to Maven 3.5.4, all of our raises have been
intermediaries. Reasoning:
* Maven 3.5.4 is almost 4 old, it is settled
* Everything before we shouls consider ancient, if you haven't moved
yet, you will not and it will be your problem.
* It is the first version which uses Maven Resolver [1]
* It will free resources to test with previous releases which most don't
do anyway
* Maybe this will make it even easier to migrate to the upcoming Maven
Core API
WDYT?
Michael
PS: I don't even mind 3.6.3 personally, thought I don't this benefit.
[1]
https://cwiki.apache.org/confluence/display/MAVEN/Maven+Ecosystem+Cleanup#MavenEcosystemCleanup-ResolverandMaven
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
